{"id":"https://openalex.org/W4400877433","doi":"https://doi.org/10.1109/tpds.2024.3431910","title":"DeepTM: Efficient Tensor Management in Heterogeneous Memory for DNN Training","display_name":"DeepTM: Efficient Tensor Management in Heterogeneous Memory for DNN Training","publication_year":2024,"publication_date":"2024-07-22","ids":{"openalex":"https://openalex.org/W4400877433","doi":"https://doi.org/10.1109/tpds.2024.3431910"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2024.3431910","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2024.3431910","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058977772","display_name":"Haoran Zhou","orcid":"https://orcid.org/0000-0001-5516-5214"},"institutions":[{"id":"https://openalex.org/I6469544","display_name":"City University of Macau","ror":"https://ror.org/04gpd4q15","country_code":"MO","type":"education","lineage":["https://openalex.org/I6469544"]},{"id":"https://openalex.org/I204512498","display_name":"University of Macau","ror":"https://ror.org/01r4q9n85","country_code":"MO","type":"education","lineage":["https://openalex.org/I204512498"]},{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN","MO"],"is_corresponding":true,"raw_author_name":"Haoran Zhou","raw_affiliation_strings":["School of Computer Science, Wuhan University, Wuhan, Hubei, China","Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science, Wuhan University, Wuhan, Hubei, China","institution_ids":["https://openalex.org/I37461747"]},{"raw_affiliation_string":"Laboratory of Internet of Things for Smart City, University of Macau, Macau, China","institution_ids":["https://openalex.org/I6469544","https://openalex.org/I204512498"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090960389","display_name":"Wei Rang","orcid":"https://orcid.org/0000-0003-4138-4867"},"institutions":[{"id":"https://openalex.org/I28006308","display_name":"Shandong Normal University","ror":"https://ror.org/01wy3h363","country_code":"CN","type":"education","lineage":["https://openalex.org/I28006308"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Rang","raw_affiliation_strings":["School of information science and engineering, Shandong Normal University, Jinan, Shandong, China"],"affiliations":[{"raw_affiliation_string":"School of information science and engineering, Shandong Normal University, Jinan, Shandong, China","institution_ids":["https://openalex.org/I28006308"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008473103","display_name":"Hongyang Chen","orcid":"https://orcid.org/0000-0002-7626-0162"},"institutions":[{"id":"https://openalex.org/I4210123185","display_name":"Zhejiang Lab","ror":"https://ror.org/02m2h7991","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210123185"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongyang Chen","raw_affiliation_strings":["Zhejiang Lab, Hangzhou, Zhejiang, China"],"affiliations":[{"raw_affiliation_string":"Zhejiang Lab, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I4210123185"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101677080","display_name":"Xiaobo Zhou","orcid":"https://orcid.org/0009-0004-9500-3390"},"institutions":[{"id":"https://openalex.org/I6469544","display_name":"City University of Macau","ror":"https://ror.org/04gpd4q15","country_code":"MO","type":"education","lineage":["https://openalex.org/I6469544"]},{"id":"https://openalex.org/I204512498","display_name":"University of Macau","ror":"https://ror.org/01r4q9n85","country_code":"MO","type":"education","lineage":["https://openalex.org/I204512498"]}],"countries":["MO"],"is_corresponding":false,"raw_author_name":"Xiaobo Zhou","raw_affiliation_strings":["Laboratory of Internet of Things for Smart City and Department of Computer and Information Science, University of Macau, Macau, China"],"affiliations":[{"raw_affiliation_string":"Laboratory of Internet of Things for Smart City and Department of Computer and Information Science, University of Macau, Macau, China","institution_ids":["https://openalex.org/I6469544","https://openalex.org/I204512498"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063911669","display_name":"Dazhao Cheng","orcid":"https://orcid.org/0000-0003-2869-7623"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dazhao Cheng","raw_affiliation_strings":["School of Computer Science, Wuhan University, Wuhan, Hubei, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science, Wuhan University, Wuhan, Hubei, China","institution_ids":["https://openalex.org/I37461747"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5058977772"],"corresponding_institution_ids":["https://openalex.org/I204512498","https://openalex.org/I37461747","https://openalex.org/I6469544"],"apc_list":null,"apc_paid":null,"fwci":0.5263,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.64257243,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":"35","issue":"11","first_page":"1920","last_page":"1935"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9993000030517578,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9993000030517578,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9962000250816345,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9796000123023987,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8153086304664612},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5822356343269348},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.5231935977935791},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.44881054759025574},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37162891030311584},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.353488564491272},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.2027975618839264},{"id":"https://openalex.org/keywords/overlay","display_name":"Overlay","score":0.1469528079032898}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8153086304664612},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5822356343269348},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.5231935977935791},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.44881054759025574},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37162891030311584},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.353488564491272},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2027975618839264},{"id":"https://openalex.org/C136085584","wikidata":"https://www.wikidata.org/wiki/Q910289","display_name":"Overlay","level":2,"score":0.1469528079032898},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2024.3431910","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2024.3431910","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3830727860","display_name":null,"funder_award_id":"62341410","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2102605133","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2155014525","https://openalex.org/W2194775991","https://openalex.org/W2271840356","https://openalex.org/W2489529491","https://openalex.org/W2525414919","https://openalex.org/W2784561332","https://openalex.org/W2892315936","https://openalex.org/W2901994046","https://openalex.org/W2919115771","https://openalex.org/W2921898934","https://openalex.org/W2978949425","https://openalex.org/W2981237538","https://openalex.org/W3010663313","https://openalex.org/W3010830594","https://openalex.org/W3012479151","https://openalex.org/W3012514909","https://openalex.org/W3022260960","https://openalex.org/W3095007491","https://openalex.org/W3157864729","https://openalex.org/W3177128963","https://openalex.org/W3203413568","https://openalex.org/W3205803342","https://openalex.org/W3216235925","https://openalex.org/W4242906757","https://openalex.org/W4246193833","https://openalex.org/W4285503890","https://openalex.org/W4287186706","https://openalex.org/W4287756266","https://openalex.org/W4293718192","https://openalex.org/W4295520929","https://openalex.org/W4297775537","https://openalex.org/W4301361180","https://openalex.org/W4318541551","https://openalex.org/W4387321556","https://openalex.org/W6637373629","https://openalex.org/W6694517276","https://openalex.org/W6732520560","https://openalex.org/W6737664043","https://openalex.org/W6748088138","https://openalex.org/W6759932407","https://openalex.org/W6762257222","https://openalex.org/W6766537505","https://openalex.org/W6779999780","https://openalex.org/W6784740996","https://openalex.org/W6790981935","https://openalex.org/W6798298412","https://openalex.org/W6857182763","https://openalex.org/W7004791202"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W2495260952","https://openalex.org/W4366179611","https://openalex.org/W2996078371"],"abstract_inverted_index":{"Deep":[0,183],"Neural":[1],"Networks":[2],"(DNNs)":[3],"have":[4],"gained":[5],"widespread":[6],"adoption":[7],"in":[8],"diverse":[9],"fields,":[10],"including":[11],"image":[12],"classification,":[13],"object":[14],"detection,":[15],"and":[16,82,120,123,139,147,176,195,222,233,245],"natural":[17],"language":[18],"processing.":[19],"However,":[20,63],"training":[21],"large-scale":[22],"DNN":[23,61,109],"models":[24],"often":[25,72],"encounters":[26],"significant":[27],"memory":[28,39,44,65,70,103,106,130,207,229],"bottlenecks,":[29],"which":[30,41],"ask":[31],"for":[32,101],"efficient":[33,95],"management":[34,58,66,97,230],"of":[35,85,135,156,171,190,218],"extensive":[36],"tensors.":[37],"Heterogeneous":[38],"system,":[40],"combines":[42],"persistent":[43],"(PM)":[45],"modules":[46],"with":[47],"traditional":[48],"DRAM,":[49],"offers":[50],"an":[51,94,133],"economically":[52],"viable":[53],"solution":[54,238],"to":[55,74,104,116,128,220,225,248,253],"address":[56],"tensor":[57,96,114,136,163,174],"challenges":[59],"during":[60,108],"training.":[62,110],"existing":[64],"methods":[67],"on":[68,203],"heterogeneous":[69,102,206],"systems":[71],"lead":[73],"low":[75,79],"PM":[76,118],"access":[77,137],"efficiency,":[78],"bandwidth":[80],"utilization,":[81],"incomplete":[83],"analysis":[84,134],"model":[86,140],"characteristics.":[87],"To":[88,186],"overcome":[89],"these":[90],"hurdles,":[91],"we":[92,142,161,193],"introduce":[93],"approach,":[98,192],"DeepTM,":[99,197],"tailored":[100],"alleviate":[105],"bottlenecks":[107],"DeepTM":[111,214],"employs":[112],"page-level":[113],"aggregation":[115],"enhance":[117],"read":[119],"write":[121],"performance":[122,146,150,216],"executes":[124],"contiguous":[125],"page":[126],"migration":[127],"increase":[129],"bandwidth.":[131],"Through":[132],"patterns":[138],"characteristics,":[141],"quantify":[143],"the":[144,149,154,169,188,199,226,240],"overall":[145],"transform":[148],"optimization":[151,180],"problem":[152],"into":[153],"framework":[155,201],"Integer":[157],"Linear":[158],"Programming.":[159],"Additionally,":[160],"achieve":[162],"heat":[164],"recognition":[165],"by":[166,242],"dynamically":[167],"adjusting":[168],"weights":[170],"four":[172],"key":[173],"characteristics":[175],"develop":[177],"a":[178,204],"global":[179],"strategy":[181],"using":[182],"Reinforcement":[184],"Learning.":[185],"validate":[187],"efficacy":[189],"our":[191,237],"implement":[194],"evaluate":[196],"utilizing":[198],"TensorFlow":[200],"running":[202],"PM-based":[205],"system.":[208],"The":[209],"experimental":[210],"results":[211],"demonstrate":[212],"that":[213],"achieves":[215,246],"improvements":[217],"up":[219,247],"36%":[221],"49%":[223],"compared":[224,252],"current":[227],"state-of-the-art":[228],"strategies":[231],"AutoTM":[232],"Sentinel,":[234],"respectively.":[235],"Furthermore,":[236],"reduces":[239],"overhead":[241],"18":[243],"times":[244],"29%":[249],"cost":[250],"reduction":[251],"AutoTM.":[254]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-02-25T08:12:03.925757","created_date":"2025-10-10T00:00:00"}